Evaluation of Sudden Cardiac Arrest by Race/Ethnicity Among Residents of Ventura County, California, 2015-2020

IMPORTANCE Sudden cardiac arrest (SCA) is a major public health problem. Owing to a lack of population-based studies in multiracial/multiethnic communities, little information is available regarding race/ethnicity-specific epidemiologic factors of SCA. OBJECTIVE To evaluate the association of race/ethnicity with burden, outcomes, and clinical profile of individuals experiencing SCA. DESIGN, SETTING, AND PARTICIPANTS A 5-year prospective, population-based cohort study of out-of-hospital SCA was conducted from February 1, 2015, to January 31, 2020, among residents of Ventura County, California (2018 population, 848112: non-Hispanic White [White], 45.8%; Hispanic/ Latino [Hispanic], 42.4%; Asian, 7.3%; and Black, 1.7% individuals). All individuals with out-of-hospital SCA of likely cardiac cause and resuscitation attempted by emergency medical services were included. EXPOSURES Data on circumstances and outcomes of SCA from prehospital emergency medical services records and data on demographics and pre-SCA clinical history from detailed archived medical records, death certificates, and autopsies. MAIN OUTCOMES AND MEASURES comorbidity RESULTS A total of 1624 patients with (1059 men; [16.1]years).Race/ethnicitydatawereavailablefor1542(95.0%)individuals,ofwhom1022(66.3%) CONCLUSIONS AND RELEVANCE The results of this study suggest that the burden of SCA was similar in Hispanic and White individuals and lower in Asian individuals. The Asian and Hispanic populations had shared SCA risk factors, which were different from those of the White population. These findings underscore the need for an improved understanding of race/ethnicity-specific differences in SCA risk.


Introduction
Out-of-hospital sudden cardiac arrest (SCA), a sudden loss of the pulse, affects approximately 350 000 individuals in the US annually. 1 Despite resuscitation efforts, the mortality rate exceeds 90%, thereby making prediction and prevention of this condition a major priority. 1 The burden of SCA is affected by multiple factors, including race, socioeconomic status (SES), and potentially race/ ethnicity. For example, population-based studies of the US Black population consistently report a 2-fold higher incidence rate compared with their White counterparts, [2][3][4][5] and there is an association between low SES and increased SCA annual incidence. 6,7 The Hispanic/Latino population is the largest and most rapidly growing racial/ethnic minority group in the US, but, to our knowledge, there have been no prospective studies evaluating SCA burden for this prominent population.
Studies of sudden cardiac death (SCD) in the US performed more than 2 decades ago suggested that the burden of this condition was significantly lower in Hispanic/Latino and Asian individuals compared with non-Hispanic White and Black individuals. 4,8 However, these retrospective studies identified SCD from death certificates, which is a relatively inaccurate method yielding a positive predictive value in the range of 19%. 9 Also, identification of SCA survivors, constituting approximately 10% of all cases, needs prospective ascertainment. Contemporary data from Korea 10 and Japan 11 suggest that rates among Asian individuals in those countries are lower than in the US overall, but data on SCA incidence among the Asian American population are lacking. Therefore, a contemporary multiethnic assessment of SCA is warranted. We conducted a prospective cohort study of SCA among all residents of Ventura County, California, a US community in which more than 40% of the residents are Hispanic.
Methods the institutional review boards of Ventura County Medical Center, Cedars-Sinai Health System, and all participating hospitals and health systems. All survivors of cardiac arrest provided written informed consent; for nonsurvivors, this requirement was waived. This study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline for cohort studies. 13 Race and ethnicity data for individual patients were obtained from death certificates, medical examiner reports, medical records, and EMS prehospital care reports. US Census data were used for the race/ethnic composition of the community. Race and ethnicity were categorized as non-Hispanic White (White), Black/African American (Black), Hispanic/Latino (Hispanic), Asian, and other. Hispanic included any individual reporting Hispanic ethnicity, regardless of race. Other included American Indian/Alaska Native and Hawaiian/Pacific Islander. Demographic characteristics and circumstances related to the cardiac arrest event were obtained from the EMS prehospital care report and characterized using the Utstein template. 14 To determine the outcome of SCA for each patient, we reviewed the EMS record and/or medical examiner report/death certificate, if applicable, and all available hospital records. Socioeconomic status was evaluated by 2 indicators: median household income and educational level. Each patient was assigned the median household income of their residential census tract, based on US Census 2018 estimates for Ventura County, and analyzed as quartiles. Individual educational achievement was obtained from the death certificate for individuals who did not survive and from an in-person interview for survivors. The clinical profile of patients was assembled from medical records, as described in detail previously. 9,12,15 Comorbidities and history of cardiovascular events were determined from physician-noted health history in medical records.
Results from autopsies were obtained from the Ventura County medical examiner.

Statistical Analysis
For calculation of crude SCA rates, incidence counts based on age and race/ethnicity were used as the numerator, and US Census 2018 5-year estimates of the Ventura County population by age and race/ ethnicity were used as the denominator. Age-adjusted incidence was calculated using US Census 2015 estimates of the standard population within each age and race/ethnicity category, using SAS, version 9.4 (SAS Institute Corp) general linear models. Comparisons of resuscitation and pre-SCA clinical profile were limited to Asian, Hispanic, and White patients with SCA owing to the small numbers in the other groups. The comparisons were performed using logistic regression models with race/ethnicity as the categorical independent variable, adjusted for age as a continuous variable to evaluate the age-adjusted association of race/ethnicity with each arrest circumstance or clinical finding. For continuous variables (age and response time in minutes), an analogous approach with analysis of variance was used, with Tukey post hoc tests for pairwise comparisons by race/ethnicity.
In addition, we evaluated sex-specific differences in clinical profile among Hispanic and White individuals using χ 2 tests. A 2-sided, unpaired value of P < .05 was considered significant. individuals was approximately 6 years younger than in White (72.6 [14.5] years) and Asian (73.6 [16.1] years) individuals (P < .001). Figure 1 illustrates differences in the age distribution by race and     . Age-specific rates in each race and ethnic group (Figure 2) show that at most ages, White and Hispanic individuals had similar rates, Black individuals had somewhat higher rates, and

JAMA Network Open | Cardiology
Asian individuals had somewhat lower rates.
Sudden cardiac arrest circumstances and outcomes overall were similar across race/ethnicity ( Table 1). Proportions with witnessed SCA and receipt of bystander cardiopulmonary resuscitation (CPR), as well as EMS response time, were comparable between the Asian, Hispanic, and White individuals with SCA. Asian individuals were 2.5 to 3 times more likely than White individuals to present with a nonshockable (pulseless electrical activity/asystole) than a shockable (ventricular fibrillation/tachycardia) rhythm; Hispanic individuals were somewhat more likely than White individuals to present with asystole. All 3 groups were similarly likely to have return of spontaneous circulation following resuscitation and to survive to hospital discharge (11.8% Asian, 13.9% Hispanic, and 13.0% White individuals; P = .69) ( Table 1). Similar survival outcomes by race/ethnicity were also observed when stratified by presenting rhythm (Table 1).
Hispanic individuals were more likely than White individuals to live in a census tract with belowmedian household income (72% vs 46%; P < .001) and less likely to have completed education beyond high school (11% vs 39%; P < .001). Income and educational levels in the Asian and White populations were similar (P Ն .12).  When stratified by sex, the overall ethnic differences in diabetes and chronic kidney disease (higher in Hispanics) and COPD and atrial fibrillation (lower in Hispanics) were observed in both men and women (Figure 3). However, other ethnic differences in risk factors were sex specific. Among

Discussion
The lack of contemporary, prospective, population-based studies of SCA in multiethnic communities represents a substantial knowledge gap in the field. These findings from the ongoing PRESTO study much lower in Hispanic individuals (men, 75 per 100 000; women, 35 per 100 000) compared with White individuals (men, 166 per 100 000; women, 74 per 100 000). 8 Similarly, based on US vital statistics mortality data from 1989 to 1998, the age-adjusted rate of SCD among the Hispanic population was just over half the rate among non-Hispanic groups. 4 However, SCA ascertainment in these previously published studies was not prospective, relied on death certificate data, and was conducted at least 2 decades ago-all factors that may explain the higher SCD rates compared with the present study. 9 A more recent study conducted in San Francisco in 2012 included all patients with SCD evaluated by the medical examiner; only those with arrhythmic causes were included in rate calculations. 16 Although the study had a small number of Hispanic individuals (21 cases of SCA, 11 with arrhythmic causes), their results were similar to ours: the annual rate of arrhythmic SCD in Hispanic individuals (13.5 per 100 000) was not significantly different (P = .40) than the rate in White individuals (20.1 per 100 000).
Our study found lower age-adjusted SCA rates among Asian compared with White individuals.
These findings are consistent with reports of substantially lower SCA incidence from studies in Korea 10 and Japan. 11 However, results from the San Francisco study indicated that Asian residents of San Francisco (18.9 per 100 000) had an annual SCA rate similar to that of White residents (P = .90). 16 In the present study, 3 established, independent factors associated with risk of SCA (ie, diabetes, hypertension, and chronic kidney disease) were significantly more common among Asian and Hispanic individuals with SCA compared with White individuals. Hispanic individuals were also more likely to have hyperlipidemia and stroke. Among Hispanic women, the differences in hypertension, hyperlipidemia, and history of stroke were more pronounced than among men, and Hispanic women were also more likely to be obese compared with their White counterparts. Despite the higher cardiovascular risk profile in the Asian and Hispanic cohorts, all groups had a similar prevalence of recognized coronary artery disease (CAD) diagnosed prior to SCA. These findings suggest race/ethnicity-specific areas of focus for education and preventive interventions that may improve screening and early diagnosis to reduce SCA burden. The finding of a lower prevalence of   21 There are dietary patterns that could be unique to Hispanic/ Latino individuals, including a higher intake of legumes and fruit, that contribute to reduced oxidative stress and benefits for endothelial function. 22,23 Culturally, a higher prevalence of social and familial support, 24 resulting in reduction of psychosocial stressors/triggers for SCA, could also contribute to protection in Hispanic/Latino individuals. 25 These as well as other yet unidentified factors warrant further detailed investigation. Furthermore, although incidence rates were equivalent to those noted in the White population, the burden of SCA among Hispanic/Latino individuals still needs to be addressed with improved methods of prediction and prevention. The differences in SCA risk factors suggest that targeted, race/ethnicity-specific reduction of risk factors could further reduce SCA burden.

JAMA Network Open | Cardiology
Our findings related to equivalence of SCA resuscitation outcomes between 3 racial/ethnic groups are also important from both public health and societal perspectives. Key factors that influence survival following resuscitation, such as response time, bystander CPR, and presenting rhythm during SCA, were similar across the groups, resulting in similar survival outcomes. Although these findings may not be generalizable, they provide data suggesting a positive role of community  32 with different patterns of migration and acculturation, all of which may affect health behaviors, risk, and outcomes. 33 In addition, there is substantial genetic diversity among Hispanic groups based on region of origin, which may also be associated with risk of SCA. 34 Therefore, our results may not be generalizable to Hispanic/Latino populations living in other regions of the US. Only 2 Hispanic individuals identified as of Black race, precluding analysis of potential differences between Black and non-Black Hispanic cohorts. The Asian population in Ventura County is relatively small and may not be representative of Asian groups in other areas of the US. In addition, the number of patients with SCA who were of Black or other race was small and precluded detailed evaluation of their clinical profile before SCA; owing to small numbers, incidence rates calculated for these groups should be interpreted with caution.

Conclusions
In this prospective multiethnic investigation of SCA in a single large US community, we observed equivalent age-adjusted incidence rates of SCA in Hispanic, White, and Black individuals and lower rates of SCA among Asian individuals. Survival outcomes following resuscitation were equivalent in the Asian, Hispanic, and White groups, despite lower socioeconomic status in Hispanic individuals.
However, the clinical risk profile of Asian and Hispanic individuals was significantly different from that of their White counterparts, and sex-specific differences were also identified. These findings highlight the need to further explore ethnicity-specific differences in SCA risk.